Introduction

How humanitarians estimate travel time

Travel time is a more prominent indicator now due to JIAF

Travel time is one small component of service access

Advances in health facility data, DTM and OSM

Computational approaches to service access

Building upon previous work with Manon

Objectives

Compare service access measurement via KI and computational approaches
Utilize computational approaches to quality check key informant data.
Identify priority locations to investigate barriers other than travel time
Reproducible approach that can be systematized

Methodology

Overview of sites, facilities and combined
Focus on 1 country Mozambique
Get IDP location data, health facility data

Analysis

Mandruzi IDP site

Map of IDP site, health sites and route to nearest facility. Comparing to KI

Figure 1: The closest health facility to Mandruzi is Centro de Saude de Dondo

Figure 2: Mandruzi is approx. 4km (49mins walk) to Centro de Saude de Dondo (browse and measure the route in the interactive map)

Spatial distribution of IDP sites and Health Facilities

Browse all IDP sites, health facilities and routes to the nearest health facility

Distribution of computed travel times

histogram of computed times…

histogram of KI times…

Comparison against KI responses

.. how many sites have different computed times to KI times?

Patterns of variance

.. what are the potential explanations as to why this variance exists?

QC - enumerators

… do some enumerators have more responses that differ from the computed travel time than others? (flagging potential issues for better training or quality control)

Spatial patterns

is there any spatial clustering of sites with a mismatch? This could be a flag to highlight areas where barriers other than travel time maybe be a key factor list of sites to further examine barriers

Findings

Ask travel time in minutes, with inclusion of mode. Check for patterns by KI which may indicate QC issues.

Limitations & next steps

Road networks Types Elevation Systematizing